All the contents disclosed in Japanese Patent Application No. 2000-000867(filed on Jan. 6, 2000), including specification, claims, drawings and abstract and summary are incorporated herein by reference in its entirety.
1. Field of the Invention
This invention relates to a database management system, and, more particularly, to the system capable of relaxing constraints on the structure of data stored in the database.
2. Description of the Related Art
In general, there are multiple records in a database management system and each record consists of multiple fields. Data structure (such as data format, data length and so on) of such fields is predetermined by the structure of the database. Thus, a high-speed data search can be carried out on the fields of the previously defined structures. Recently, extensible markup language (XML) data is becoming popular as structure descriptive data. Data stored in a system can be attributed by the XML information so as to indicate which fields it belongs to, i.e., so that the hierarchical structure can be automatically extracted. For example, a computer stores an item xe2x80x9cIchiro YAMADAxe2x80x9d, and another item xe2x80x9cmanagerxe2x80x9d respectively into fields xe2x80x9cnamexe2x80x9d and xe2x80x9cpositionxe2x80x9d by referring to tags in each item given XML data shown in FIG. 3A.
The database management system, however, has the following problems to be solved. If the record length is fixed, both the number and the length of fields in each record are fixed. Therefore, the data structures need to be rebuilt when new fields are added and/or the field lengths are varied. In other words, data contained in added attributes is discard even when XML attributes containing the new data are provided. Usually, a database management system manages a huge amount of data so that many extremely complicated processes need to be carried out for restructuring the entire database even if it is for just one record out of hundreds of thousands of records managed by the system.
Furthermore, data (such as null) adjusted to the new data structure needs to be entered even when no data exists for the field so that this causes unused regions therein.
In order to solve such problems, even if each of the records is defined to have variable length, the length of each field may be variable, but the number of field(s) in each of the records must be predefined. As a consequence, no additional field can be added.
In other words, the conventional database management systems do not employ a structure sufficiently supporting the advantages of XML data by which addition of attributes to the data itself can be performed. Although, XML data can describe a relational structure of each item using the tag described above, such relational structure can not be stored as it is by the conventional database management system.
It is an object of the present invention to solve the above-mentioned problems and to provide a database management system having sufficient flexibility to make use of the advantages of descriptively attributed data.
1) In accordance with one aspect, the present invention provides a database management system comprising:
data group storage means for storing multiple groups of data including attribution items having predetermined structure, without destroying the structure of the data stored;
auxiliary data storage means for storing auxiliary index data for the groups of data with attribution items, showing the location of the data for search purposes; and
append data group processing means for appending newly provided data, appending the groups of data to the stored data, appending the relevant location data for multiple attribution items when auxiliary search data already exists matching the attribution items of the newly appended groups of data, but appending the data of the newly provided attribution items as well as relevant location data when matching auxiliary search data does not exist.
4) In accordance with another aspect, the present invention provides a database management system comprising:
data group storage means for storing multiple groups of data including attribution items having predetermined structure, without destroying the structure of the data stored;
auxiliary data storage means for storing auxiliary data for said groups of data with attribution items, showing the location of the data for search purposes; and
searching means for searching data stored by the data group storage means with reference to auxiliary data for search stored by the auxiliary data storage means when a search command including attribution items to be searched is provided.
6) In accordance with another aspect, the present invention provides a database management system using a computer for managing XML block data including optionally multiple attribution items, the structure of each of the attribution items being formed according to a tree structure,
wherein the system controls the storing process so that the XML block data is stored into a block of primary storage region per each of the XML block data while maintaining the structure of the XML block data, and the system controls the storing process so that location data for locating the stored XML block data is stored into a secondary storage region, the location data showing the location of the XML block data for search purposes, and
wherein the following processing is performed when new XML block data is entered;
a) appending the new XML block data into the primary storage region while maintaining its tree structure, and
b) among the attribution items of the newly provided XML block data, for those which match the properties of the index data, the location information of the attribution items is stored in the secondary storage region, and for those which do not have attribution items in the stored index data, new attribution items are stored in the secondary storage region along with the location data for the new attribution items.
8) In accordance with another aspect, the present invention provides a method of managing data using a computer comprising the steps of:
storing in a primary storage region, multiple groups of data including attribution items having predetermined structure, without destroying the structure of the data stored;
storing in a secondary storage region, auxiliary data for the data groups with attribution items, showing the location of the data for search purposes; and
accepting into the primary storage region, newly entered groups of data with attribution items, appending the newly entered data groups, matching attribution items from the data groups, appending auxiliary data for the attribution items, appending location data for search purposes, and appending auxiliary data for attribution items not present.
9) In accordance with another aspect, the present invention provides a method of searching data using a computer comprising the steps of:
storing in a primary storage region, multiple groups of data including attribution items having predetermined structure, without destroying the structure of the data stored;
storing in a secondary storage region, auxiliary data for the data groups with attribution items, showing the location of the data for search purposes; and
searching multiple groups of data stored in the primary storage region with reference to auxiliary search data when a search command including attribution items to be searched is provided.
12) In accordance with another aspect, the present invention provides a data storing medium including a data structure comprising the following regions:
A) a data group storage region for storing multiple groups of data including attribution items having predetermined structure, without destroying the structure of the data stored; and
B) an auxiliary data storage region for storing auxiliary data for the groups of data with attribution items, showing the location of the data for search purposes.
The term xe2x80x9cattribution itemxe2x80x9d is generally referred to in the literature as xe2x80x9cattributexe2x80x9d or xe2x80x9cattributionxe2x80x9d indicating an explanatory data item, or property, or other data item appended to the principle data item.